1 citations · 1 across the 3 of their papers we have counts for
4 papers
Training Trajectories Determine Circuit Removability in Annealable Soft-Prior Transformers
Zonglin Yang, Ziming Zhao, Wei Tang +5
Soft positional priors can help small Transformers learn retrieval circuits, but it is unclear whether the resulting circuits remain functional once the prior is removed. We test t…
UOT-Gap: A Variational Principle for the Modality Gap in Vision-Language Models via Unbalanced Optimal Transport
Zonglin Yang, Huilan Ma, Xudan Zheng +1
Vision-language models such as CLIP embed images and text in a shared space, where modality-specific distributions often remain separated. Existing accounts connect this modality g…
SciIntegrity-Bench: A Benchmark for Evaluating Academic Integrity in AI Scientist Systems
Zonglin Yang, Xingtong Liu, Xinyan Xu
AI scientist systems are increasingly deployed for autonomous research, yet their academic integrity has never been systematically evaluated. We introduce SCIINTEGRITY-BENCH, the f…
Efficient and provably convergent end-to-end training of deep neural networks with linear constraints
Zonglin Yang, Zhexuan Gu, Yancheng Yuan
Training a deep neural network with the outputs of selected layers satisfying linear constraints is required in many contemporary data-driven applications. While this can be achiev…